Rolling of thin strip and foil: Application of a tribological model for "mixed" lubrication

A mechanical model of cold rolling of foil is coupled with a sophisticated tribological model. The tribological model treats the "mixed" lubrication regime of practical interest, in which there is "real" contact between the roll and strip as well as pressurized oil between the surfaces. The variation of oil film thickness and contact ratio in the bite is found by considering flattening of asperities on the foil and the build-up of hydrodynamic pressure through the bite. The boundary friction coefficient for the contact areas is taken from strip drawing tests under similar tribological conditions. Theoretical results confirm that roll load and forward slip decrease with increasing rolling speed due to the decrease in contact ratio and friction. The predictions of the model are verified using mill trials under industrial conditions. For both thin strip and foil, the load predicted by the model has reasonable agreement with the measurements. For rolling of foil, forward slip is overestimated. This is greatly improved if a variation of friction through the bite is considered.

Foot contact detection for sprint training

We introduce a new algorithm to automatically identify the time and pixel location of foot contact events in high speed video of sprinters. We use this information to autonomously synchronise and overlay multiple recorded performances to provide feedback to athletes and coaches during their training sessions. The algorithm exploits the variation in speed of different parts of the body during sprinting. We use an array of foreground accumulators to identify short-term static pixels and a temporal analysis of the associated static regions to identify foot contacts. We evaluated the technique using 13 videos of three sprinters. It successfully identifed 55 of the 56 contacts, with a mean localisation error of 1.39±1.05 pixels. Some videos were also seen to produce additional, spurious contacts. We present heuristics to help identify the true contacts. © 2011 Springer-Verlag Berlin Heidelberg.

The role of adaptive thermal comfort in the prediction of the thermal performance of a modern mixed-mode office building in the UK under climate change

Climate change is becoming a serious issue for the construction industry, since the time scales at which climate change takes place can be expected to show a true impact on the thermal performance of buildings and HVAC systems. In predicting this future building performance by means of building simulation, the underlying assumptions regarding thermal comfort conditions and the related heating, ventilating and air conditioning (HVAC) control set points become important. This article studies the thermal performance of a reference office building with mixedmode ventilation in the UK, using static and adaptive thermal approaches, for a series of time horizons (2020, 2050 and 2080). Results demonstrate the importance of the implementation of adaptive thermal comfort models, and underpin the case for its use in climate change impact studies. Adaptive thermal comfort can also be used by building designers to make buildings more resilient towards change. © 2010 International Building Performance Simulation Association (IBPSA).

Perceptual learning, roving and the unsupervised bias

Perceptual learning improves perception through training. Perceptual learning improves with most stimulus types but fails when . certain stimulus types are mixed during training (roving). This result is surprising because classical supervised and unsupervised neural network models can cope easily with roving conditions. What makes humans so inferior compared to these models? As experimental and conceptual work has shown, human perceptual learning is neither supervised nor unsupervised but reward-based learning. Reward-based learning suffers from the so-called unsupervised bias, i.e., to prevent synaptic " drift" , the . average reward has to be exactly estimated. However, this is impossible when two or more stimulus types with different rewards are presented during training (and the reward is estimated by a running average). For this reason, we propose no learning occurs in roving conditions. However, roving hinders perceptual learning only for combinations of similar stimulus types but not for dissimilar ones. In this latter case, we propose that a critic can estimate the reward for each stimulus type separately. One implication of our analysis is that the critic cannot be located in the visual system. © 2011 Elsevier Ltd.

Gaussian Process training with input noise

In standard Gaussian Process regression input locations are assumed to be noise free. We present a simple yet effective GP model for training on input points corrupted by i.i.d. Gaussian noise. To make computations tractable we use a local linear expansion about each input point. This allows the input noise to be recast as output noise proportional to the squared gradient of the GP posterior mean. The input noise variances are inferred from the data as extra hyperparameters. They are trained alongside other hyperparameters by the usual method of maximisation of the marginal likelihood. Training uses an iterative scheme, which alternates between optimising the hyperparameters and calculating the posterior gradient. Analytic predictive moments can then be found for Gaussian distributed test points. We compare our model to others over a range of different regression problems and show that it improves over current methods.

PH-dependent leaching behaviour and other performance properties of cement-treated mixed contaminated soil

Portland cement has been widely used for stabilisation/solidification (S/S) treatment of contaminated soils. However, there is a dearth of literature on pH-dependent leaching of contaminants from cement-treated soils. This study investigates the leachability of Cu, Pb, Ni, Zn and total petroleum hydrocarbons (TPH) from a mixed contaminated soil. A sandy soil was spiked with 3000 mg/kg each of Cd, Cu, Pb, Ni and Zn, and 10,000 mg/kg of diesel, and treated with ordinary Portland cement (CEM I). Four different binder dosages, 5%, 10%, 15% and 20% (m/m) and different water contents ranging from 13%-19% dry weight were used in order to find a safe operating envelope for the treatment process. The pH-dependent leaching behaviour of the treated soil was monitored over an 84-day period using a 3-point acid neutralisation capacity (ANC) test. The monolithic leaching test was also conducted. Geotechnical properties such as unconfined compressive strength (UCS), hydraulic conductivity and porosity were assessed over time. The treated soils recorded lower leachate concentrations of Ni and Zn compared to the untreated soil at the same pH depending on binder dosage. The binder had problems with Pb stabilisation and TPH leachability was independent of pH and binder dosage. The hydraulic conductivity of the mixes was generally of the order, 10-8 m/sec, while the porosity ranged from 26%-44%. The results of selected performance properties are compared with regulatory limits and the range of operating variables that lead to acceptable performance described. © 2012 The Research Centre for Eco-Environmental Sciences, Chinese Academy of Sciences.

Factor analysis based VTS discriminative adaptive training

Vector Taylor Series (VTS) model based compensation is a powerful approach for noise robust speech recognition. An important extension to this approach is VTS adaptive training (VAT), which allows canonical models to be estimated on diverse noise-degraded training data. These canonical model can be estimated using EM-based approaches, allowing simple extensions to discriminative VAT (DVAT). However to ensure a diagonal corrupted speech covariance matrix the Jacobian (loading matrix) relating the noise and clean speech is diagonalised. In this work an approach for yielding optimal diagonal loading matrices based on minimising the expected KL-divergence between the diagonal loading matrix and "correct" distributions is proposed. The performance of DVAT using the standard and optimal diagonalisation was evaluated on both in-car collected data and the Aurora4 task. © 2012 IEEE.